THE JOURNAL OF BIO~~ICAL CHEMISTRY Vol. 237, No. 5, May 1962 Printed in U.S.A

The Reaction Sequence in Ergothioneine Biosynthesis: Hercynine as an Intermediate*

AMIR ASKARI AND DONALD B. MELVILLE

From the Departments of Biochemistry, Cornell University Medical College, New York, New York, and the University of Vermont Collegeof Medicine, Burlington, Vermont

(Received for publication, January 9, 1962) Downloaded from

Ergothioneine can be synthesized by the mold, Neurospora provide data which show that hercynine is an intermediate com- crassa, from the amino acids, , methionine, and cysteine pound in the biosynthesis of ergothioneine. (1). The intact histidine molecule is used in the synthesis, and EXPERIMENTAL PROCEDURE the three methyl groups of the betaine moiety can be derived http://www.jbc.org/ from methionine by a triple transmethylation process (2). Cys- Materiak-L -Histidine-2 - Cl4 dihydrochloride was prepared teine appears to be an immediate precursor of the atom from NaC14N (4, 5). Sodium formate- was purchased from of the thiolimidazole ring. In this paper we present data which Tracerlab, Inc., Boston, Massachusetts. Ergothioneine was pur- establish the sequence of these biosynthetic reactions. chased from Nutritional Biochemicals Corporation, Cleveland, Of the several possible synthetic routes, the most straight- Ohio, and was recrystallized from aqueous ethanol. Methods-N. crassa (ATCC 10336, wild type A) was grown in forward involve as intermediates either thiolhistidine or hercy- at UNIVERSITY OF TOLEDO LIBRARIES on March 24, 2016 nine, the betaine of histidine, according to the depicted reaction a chemically defined medium (6) in shaken cultures as previously schemes (Scheme 1). described (1). Ergothioneine was extracted from the dried CH===C-CH2rCH-COOH CH=C-CH+H-COOH I k NH NH 2 Cyst&e N AH I&H* ’ NC/ I SH Histidine Thiolhistidine

ethionine ethionine i? i? CH=C-CHe--CH-COO- CH=+CH+H-COO-

Hercynine Ergothioneine SCHEME 1 The sequence of reactions involving thiolhistidine as an inter- mycelium with hot water and was purified by chromatography mediate has not been substantiated by studies which have been on alumina (7, 8). Column fractions were analyzed for ergo- done with thiolhistidine-2-P (1). The alternative pathway thioneine by the modified Hunter diazo reaction (9). Dried through hercynine has not been explored heretofore. Although aliquots in steel planchets were examined for radioactivity with hercynine was identified several years ago as a constituent of a mica window counter. mushrooms (3), the substance has not been well characterized, Preparation of Hercynine-Hercynine was prepared by the and no evidence for its biological relationship to ergothioneine oxidation of ergothioneine with ferric chloride (10). A solution has been adduced. In this paper we describe the preparation of of 100 mg of ergothioneine in 0.5 ml of water was mixed with 672 crystalline hercynine, including the CWlabeled compound, and mg of FeC13.6Hz0 dissolved in 6 ml of water. The mixture was heated under reflux for 30 minutes and then was cooled to room * This work was aided by grants from the National Science temperature. An aqueous 10% solution of Na2C03 was added, Foundation and the National Institute of Allergy and Infectious a few drops at a time with centrifugation, until the supernatant Diseases, United States Public Health Service. Requests for reprints should be addressed to Dr. Donald B. Melville, Depart- solution showed no further precipitation on the addition of ment of Biochemistry, University of Vermont College of Medicine, Na2C03. The final supernatant solution was removed, and the Burlington, Vermont. precipitate was washed twice with 3-ml portions of water. The

1615 1616 The Reaction Sequence in Ergothioneine Biosynthesis Vol. 237, No. 5 combined solutions were evaporated in a vacuum to dryness. paper chromatography with n-butanol-methanol-benzene-water The residue was extracted with three 4-ml portions of hot metha- (1:2:1:1) as the solvent. nol, and the combined extracts were filtered, made acid to lit- A 75-pg sample of the product, prepared as a thin film for mus with a few drops of concentrated HCl, refiltered, and evapo- measurement of radioactivity, showed 2180 c.p.m. A 150.pg rated to dryness in a vacuum. The residue was extracted with sample was added to an aqueous solution of 5 mg of nonisotopic three 2-ml portions of dimethylformamide, and the extracts were ergothioneine and 5 mg of ISa$04, and the mixture was treated combined, filtered, and evaporated to dryness in a vacuum. with an excess of bromine-water. The addition of barium chlo- This residue was chromatographed on a column of 20 g of alu- ride yielded barium sulfate, which was collected as a pad on filter mina with 100 ml of 80% ethanol as the solvent, 5-ml fractions paper and showed 9 c.p.m. being collected. Aliquots of each were tested with the modified Biosynthesis of Ergothioneine from Doubly Labeled Hercynine- Hunter diazo test for ergothioneine. Those fractions giving a Cl-A 48-hour culture of N. crassa was washed with sterile yellow color in the test were combined, the solvent was removed water and then was suspended in a sterile solution of 6 mg of the under reduced pressure, and the residue was extracted with three hercynine-Cl4 in 100 ml of water. The mixture was shaken for

l-ml portions of methanol. Ether was added to the combined 24 hours, and the mycelium was collected, washed with water, Downloaded from filtered extracts until the solution was turbid. On standing and autoclaved with 35 ml of water containing 10 mg of non- overnight in the refrigerator, the solution deposited rosettes of isotopic ergothioneine. The aqueous extract was chromato- small rod-shaped crystals which were washed with ether and graphed on alumina with 75% ethanol-l% formic acid, and the dried. The crystals weighed 35 mg; an additional 10 mg were ergothioneine-containing fractions were combined and chromato- obtained from the mother liquors. For analysis the material graphed twice on alumina, each time with 80% ethanol as the was recrystallized twice from methanol-ether mixtures. A test solvent. Analysis of aliquots from the effluent fractions from http://www.jbc.org/ for ionic halogen was negative. the last column showed excellent correspondence between ergo- GH1502N3 (197.2) thioneine and radioactivity determinations. These fractions were combined, 20 mg of nonisotopic ergothioneine were added, Calculated: C 54.82, H 7.67, N 21.31 and the ergothioneine was crystallized from aqueous ethanol to Found : C 54.72, H 7.85, N 21.04 yield 23.6 mg of radioactive product. Recrystallization did at UNIVERSITY OF TOLEDO LIBRARIES on March 24, 2016 Hercynine prepared in this fashion is a colorless substance which not decrease the specific activity. The thrice crystallized com- is readily soluble in water, methanol, hot ethanol, acetone, and pound was used for degradation to thiolurocanic acid and tri- dimethylformamide, but is insoluble in ether, benzene, chloro- methylamine by treatment with hot concentrated NaOH as form, and ethyl acetate. It is hygroscopic, particularly when described previously (1). From 8.8 mg of the ergothioneine impure. The compound melts with decomposition at 235-237” there were obtained, after crystallization of the reaction products (corrected) on the micro melting point stage, with charring to constant specific radioactivities, 1.6 mg of thiolurocanic acid above 210”. No absorption occurs in the ultraviolet range at and 1.0 mg of trimethylamine chloroplatinate showing 302 c.p.m. wave lengths longer than 230 mp. [a]: +51.5” (c = 1.49 in and 116 c.p.m., respectively. water). In order to compare these radioactivities with that of the Preparation of Doubly Labeled Hercynine-Cl-Hercynine doubly labeled hercynine, 150 pg of the W-labeled hercynine labeled with Cl4 in both the ring and the methyl groups were mixed with 12.6 mg of nonisotopic hercynine, and the mix- was obtained by the degradation of doubly labeled ergothio- ture was subjected to treatment with hot concentrated alkali as neine-Cl4, which in turn was prepared biosynthetically by grow- described for ergothioneine. Trimethylamine was readily lib- ing N. crassa in the presence of formate-C14. To each of 10 erated under these conditions and was purified to constant flasks, each containing 100 ml of growth medium, were added radioactivity as the chloroplatinate. For the isolation of uro- 5.44 mg (0.08 mc) of sodium formate-CX4. To one of the flasks canic acid, the alkaline reaction mixture was acidified, evaporated approximately 0.25 me of carrier-free sulfuric acid-S35 was also to dryness in a vacuum, and extracted with dimethylformamide. added. N. crassa was grown in the media from spore inocula Urocanic acid was precipitated from the filtered extract by the for 96 hours. Each mycelium was then shaken for 24 hours in addition of ether and was purified to constant radioactivity by 100 ml of a sterilized aqueous solution of isotopic formate, or repeated crystallization from dimethylformamide-ether mixtures. formate and sulfate, in the same concentrations as were used in The final product was obtained as 1.8 mg of fine needles, micro the growth medium. Each mycelium was then collected on a m.p. 222-226”, with a radioactivity of 276 c.p.m. The trimeth- filter and heated in an autoclave at 15 pounds of pressure for 30 ylamine chloroplatinate weighed 4.8 mg and showed 362 c.p.m. minutes with 30 ml of water. The mycelia were washed twice The radioactivities of the doubly labeled hercynine-Cl4 and with small portions of hot water, and the combined extracts and the ergothioneineCi4 synthesized from it by N. crassa are com- washings were evaporated to dryness in a vacuum after the addi- pared in Table I. tion of 40 mg of ergothioneine. The residue was chromato- Competition between Histidine and Hercynine in Biosynthesis of graphed on alumina with 75% ethanol-l% formic acid as the Ergothioneine-The mycelia from two 4%hour growth cultures of solvent. The effluent fractions which contained ergothioneine, N. crassa were collected together on a filter and washed with as determined by the diazo test, were combined and evaporated water under sterile conditions. The mycelial pad was divided to dryness. This material was then treated with ferric chloride into two equal parts. One part was placed in a sterile solution as described above for the preparation of hercynine. From the of 3.5 mg of histidine-Cr4.2HCl (137,000 c.p.m.) in 100 ml of reaction mixture, 23 mg of twice crystallized product were ob- water. The remaining mycelium was placed in a similar solution tained which gave a single diazo-positive (11) and radioactive of 3.5 mg of histidine-Cl4 .2HCl and 9 mg of nonisotopic hercy- spot with the same RF as that shown by hercynine (0.46) after nine. Each was incubated at 25-27” for 24 hours on a rotary May 1962 A. Askari and D. B. Melville 1617

shaker. Hot water extracts were chromatographed on alumina, It was a consideration of these alternative possibilities which first with 75% aqueous ethanol containing 1% formic acid, and dictated the use of hercynine-04 labeled both in the methyl then with 80% aqueous ethanol, and the effluent fractions were groups and in the imidazole ring, since in either case a preferential analyzed for ergothioneine and radioactivity as described earlier. loss of Cl4 from one of these positions would occur if only part of The results are shown in Fig. 1. the hercynine molecule were incorporated into ergothioneine. The fractions constituting the second radioactive peak in each A comparison of the isotope distribution in the hercynine-C*4 of the last chromatograms gave the typical magenta-colored used as the precursor with the distribution in the ergothio- diazo test for ergothioneine. The fractions associated with the neine-Cl4 synthesized from it was obtained by decomposition of first radioactive peak gave yellow to orange colors in the diazo each with hot alkali, which liberates trimethylamine from both test, and these were particularly strong in the fractions from the compounds. The specific radioactivities of the trimethylamine competition experiment with hercynine and histidine-Cl4. These and urocanic acid obtained from hercynine, and of the trimethyl- latter fractions were combined, 10 mg of nonisotopic hercynine amine and thiolurocanic acid from ergothioneine, are compared were added, and the hercynine was recovered by crystallization in Table I. Within experimental error, the distribution patterns

from methanol-ether. The product weighed 8.6 mg and showed are identical for the hercynine and ergothioneine. This result Downloaded from 816 c.p.m. The specific radioactivity was unchanged after a shows beyond reasonable doubt that the intact hercynine mole- second crystallization from methanol-ether. A 7.1.mg sample cule was incorporated into ergothioneine. was decomposed with alkali as described earlier. Crystalline The results of the competition experiment, in which Neurospora urocanic acid (1.4 mg) was isolat,ed and showed 308 c.p.m. The mycelia were incubated with aqueous solutions of histidine-CY, specific activity did not decrease on recrystallization of the with and without the addition of nonisotopic hercynine, provide sample. http://www.jbc.org/

TABLE I RESULTS AND DISCUSSION Distribution of radioactivity in ergothioneine synthesized The preparation of hercynine as the crystalline free base, as from hercynine-Cl4 by Neurospora crassa described in “Experimental Procedure,” has permitted unam- Wild type N. crassa was incubated with hercynine labeled with

biguous studies of its ability to serve as a precursor for the syn- Pin both the imidazole ring and the methyl groups of the betaine at UNIVERSITY OF TOLEDO LIBRARIES on March 24, 2016 thesis of ergothioneine by N. crassa. The most convenient moiety. The resulting ergothioneine was degraded with alkali to method of preparing isotopically labeled hercynine for such thiolurocanic acid and trimethylamine. For comparison of studies is by oxidative removal of the sulfur atom from corre- radioactivities, a sample of the hercynine-Cl4 was degraded in spondingly labeled ergothioneine. The labeled ergothioneine similar fashion to urocanic acid and trimethylamine.

is most readily prepared biosynthetically. In the present work, Ratio of Compound Product acid to tri- since it was desirable to employ hercynine labeled in two posi- methylamine tions with isotope, formate was used as the precursor for the biosynthesis of the ergothioneine because it had been found earlier c.p.?%/mm01e (1) that this precursor leads to labeling in both the trimethyl- Hercynine Urocanic acid 32,100 1.05 ammonium radical and the imidazole ring of ergothioneine. Trimethylamine 30,600 To permit the detection of any ergothioneine-Cl4 which might contaminate the labeled hercynine because of incomplete oxida- Ergothioneine Thiolurocanic acid 21,100 1.06 tion of the ergothioneine, sulfate-S35 was added to the growth Trimethylamine 19,900 / medium in order to label the sulfur atom of ergothioneine. The subsequent treatment of the hercynine with bromine-water, I I I I I I I I I I which converts the sulfur of ergothioneine to inorganic sulfate d4-HISTIDINE d4-HISTIDINE 300 (lo), yielded a carrier sulfate fraction which contained negligible HERCYNINE radioactivity. This demonstrated that no significant amount of ii ergothioneine-Cl4 was present in the hercynine-C14. a Initial experiments with hercynine as a possible precursor of 0 ergothioneine yielded negative results. It was established that this was due to the impermeability of the Neurospora mycelium to hercynine. Since work in our laboratory by W. Feldman had shown that a similar impermeability of the mycelium toward ergothioneine could be overcome by incubating the mycelium in water rather than in growth medium, the same technique was tried with hercynine and proved to be successful. When doubly labeled hercynine-Cl4 was incubated in aqueous solution with Neurospora mycelium, a significant amount of Cl4 appeared in the ergothioneine which was subsequently isolated 20 40 60 SO 100 20 40 60 SO 100 from the mycelium. This fact in itself strongly indicates that EFFLUENT ML. hercynine is a precursor of ergothioneine. On the other hand, it FIG. 1. Distribution of ergothioneine (0) and radioactivity does not rule out the possibility that only the methyl groups of (0) in alumina chromatograms of extracts of Neurospora crussu, hercynine are used, or that hercynine is first demethylated to showing the effect of nonisotopic hercynine on the incorporation histidine, which is a known precursor of ergothioneine (1, 2). of histidine-Cl4 into ergothioneine. 1618 The Reaction Sequence in Ergothioneine Biosynthesis Vol. 237, No. 5 further evidence for the utilization of hercynine in the synthesis mals (5). The possibility that hercynine is formed in the catabo- of ergothioneine and in addition indicate that hercynine is a lism of ergothioneine in animal tissues has not been investigated. normal constituent of Neurospora crassa mycelium. A compari- Presently available data suggest that the occurrence of both son of the distribution of radioactivity and ergothioneine in hercynine and ergothioneine in animal tissues may well be as- chromatographic fractions of the mycelial extracts (Fig. 1) shows cribed, at least in part, to dietary constituents which contain that the amount of histidine-Cl4 converted into crgothioneine is both compounds because of a primigenial fungal origin. markedly reduced by the presence of nonisotopic hercynine in the incubation mixture, as would be expected if hercynine is an SUMMARY intermediate in the biosynthetic reaction sequence. At the same With the aid of V-labeled histidine, it has been found that time, a radioactive peak preceding the ergothioneine peak in the hercynine, the betaine of histidine, is present in Neurospora effluent from the columns was greatly increased when nonisotopic crassa mycelium and is synthesized from histidine by this . hercynine was added to the incubation mixture. The substance Hercynine has been characterized as the crystalline free base. responsible for this radioactivity was identified as hercynine by Hercynine which was doubly labeled with C?J, in both the imidaz- its chromatographic behavior, by its characteristic color in the ole ring and the trimethylamine moiety, was prepared and was Downloaded from diazo test, by cocrystallization with authentic hercynine, and by found to be converted to ergothioneine by N. crassa without conversion to radioactive urocanic acid on treatment with preferential loss of isotope, thus demonstrating that the intact alkali. The marked increase in radioactivity in hercynine hercynine molecule is incorporated into ergothioneine. The which occurred concomitantly with the decreased amount of time sequence of biosynthetic reactions involved in the bio- radioactivity in ergothioneine can be explained as the dilution

synthesis of ergothioneine has therefore been established to be http://www.jbc.org/ and “trapping” by nonisotopic hercynine of hercynine-Cl4 syn- of histidine to form hercynine, followed by sulfhy- thesized by Neurosvora from histidine-C14. dration of hercynine to form ergothioneine. The presence of hercynine in mycelial extracts can be de- tected most readily by means of the diazo test after the extracts REFERENCES have been chromatographed with aqueous ethanol as the solvent. 1. MELVILLE, D. B., EICH, S., AND LUDWIG, M. L., J. Biol. Chem., When formic acid is present in the developing solvent, hercynine 224, 871 (1957). at UNIVERSITY OF TOLEDO LIBRARIES on March 24, 2016 and ergothioneine overlap in the effluent, and the yellow or 2. MELVILLE, D. B., LUDWIG, M. L., INAMINE, E., AND RACHELE, orange color produced by hercynine is obscured by the magenta J. R., J. Biol. Chem., 234, 1195 (1959). 3. KUTSC&R, F., 2. Unterkch. Nahr..u. G&ussm., 21,535 (1911). color given by ergothioneine. A previously unidentified radio- 4. BORSOOK. H.. DEASY. C. L.. HAAGEN-SMIT. A. J.. KEIGHLEY. active peak observed earlier (1) in chromatograms prepared from G., AND LOGY, P. fi., J. k&02. Chem., 187,839 (i950). ’ N. crassa which had been grown in the presence of histidine-Cl4 5. MELVILLE, D. B., HORNER, W. H., OTBEN, C. C., AND LUD- can, on the basis of the present data, be ascribed to hercynine. WIG, M. L., J. Biol. Chem., 213, 61 (1955). Similar peaks have been observed when formateCi4 or methyl- 6. HOROWITZ, N. H.. AND BEADLE, G. W., J. Biol. Chem., 160, 325 (1943). ’ labeled methionine-Cl4 were used as precursors. 7. MELVILLE. D. B.. GENGHOF. D. S.. INAMINE. E.. AND KOVAL- The physiological significance of hercynine is unknown. This ENKO, V:, J. Biol. Chem.,‘223, 9’(1956). ’ ’ betaine was isolated from the edible mushroom, Agaricus cam- 8. MELVILLE, D. B., HORNER, W. H., AND LUBSCHEZ, R., J. pestris, as the gold salt by Kutscher (3), and from Biol. Chem., 206, 221 (1954). 9. MELVILLE, D. B., AND LUBSCHEZ, R., J. Biol. Chem., 200, 275 as the picrate and gold salt by Reuter (12). At approximately (1953). the same time, Barger and Ewins oxidized ergothioneine with 10. BARGER, G., AND EWINS, A. J., J. Chem. Sot., 2336 (1911). ferric chloride and obtained the picrate of a substance which they 11. AYES. B. N.. AND MITCHELL. H. K.. J. Am. Chem. Sot., 74, considered to be histidine betaine (10). Further work confirmed 252’(1952).’ 12. REUTER, C., 2. physiol. Chem., 78, 167 (1912). the identity of the two materials (13, 14). Syntheses of hercy- 13. BARGER, G., AND EWINS, A. J., Biochem. J., ‘7,204 (1913). nine based on the condensation of a-chloroimidazolepropionic 14. WINTERSTEIN, E., AND I&XTTE& C., 2. phys>ol. Chek., 86, 234 acid with trimethylamine have been described (15, 16), but the (1913). free base was not adequately characterized. More recently, 15. ENGELAND. R.. AND KUTSCHER, F., Zentr. Phusiol.,” 26, 569 Ackermann, List, and Menssen have obtained evidence for the (1912). ’ ’ 16. PASINI, C., AND VERCELLONE, A., Gazz. chim. ital., 86, 349 presence of hercynine in the king crab (17), in the red blood cells (1955). of cattle and the seminal fluid of the boar (18), and in a snail 17. ACKERMANN, D., AND LIST, P. H., 2. physiol. Chem., 313, 30 (19). These workers also detected ergothioneine in the king (1958). crab and confirmed its occurrence in cattle red cells and boar 18. ACHERMANN, D., LIST, P. H., AND MENSSEN, H. G., 2. physiol. Chem., 314, 33 (1959). seminal fluid. From this association, they have suggested that 19. ACHERMANN, D., AND MENSSEN, H. G., 2. physiol. Chem., 318, hercynine serves as a precursor for the biosynthesis of ergothio- 212 (1960). neine in these species. It should be pointed out that, although 20. HEATH, H., RIMINGTON, C., GLOVER, T., MANN, T., AND lower animals have not been investigated, earlier studies (20) LEONE, E., Biochem. J., 64, 606 (1953). 31 which appeared to demonstrate the synthesis of ergothioneine in -a. MELVILLE. D. B.. OTKEN. C. C.. AND KOVALENKO, V., J. Biol. Chhkm., 216; 325 (1955). ’ the pig have not been confirmed (21,22); indeed, no evidence has 22. HEATH, H., RIMINGTON, C., AND MANN, T., Biochem. J., 66, been found for the synthesis of ergothioneine in any higher ani- 369 (1957). The Reaction Sequence in Ergothioneine Biosynthesis: Hercynine as an Intermediate Amir Askari and Donald B. Melville J. Biol. Chem. 1962, 237:1615-1618.

Access the most updated version of this article at http://www.jbc.org/content/237/5/1615.citation

Downloaded from Alerts: • When this article is cited • When a correction for this article is posted

Click here to choose from all of JBC's e-mail alerts http://www.jbc.org/ This article cites 0 references, 0 of which can be accessed free at http://www.jbc.org/content/237/5/1615.citation.full.html#ref-list-1

at UNIVERSITY OF TOLEDO LIBRARIES on March 24, 2016